Not All Laws Are the Same

I'm glad to see that my internet acquaintance John Althouse Cohen is fine, despite living in southern Manhattan. In fact, his internet connection seems to be fine as well. He writes:

The traffic in the blackout areas of Manhattan is lawless in the most literal sense: the traffic lights aren't working, so the law cannot be applied as usual. But "lawless" doesn't seem to be a fitting description; the driving seems better-behaved than usual. We're so used to seeing people act under a system of government rules that it's easy to assume that without the rules, everything would descend into chaos. But perhaps free people are generally capable of acting decently on their own. Of course, that's never going to be universal; but then, people break the law too. In fact, a dense set of rules tempts people to see how close to (or how far across) the borderline of legality they can go without being penalized. In the absence of governmental laws, people might focus more on other kinds of laws: social norms and ethics. 

[Added: There actually is a law that applies when the traffic lights aren't working, but people probably don't know about that law, and they definitely aren't following it . . .]

Unsystematic evidence suggests that feelings of solidarity are particularly strong in the immediate aftermath of a common negative experience, with people on their best behaviour accordingly. Should Manhattan traffic lights continue to not work, I would expect drivers' level of care to slowly deteriorate. So perhaps one shouldn't draw too much of a conclusion from current driving behaviour to driving behaviour in the absence of traffic lights more generally. More importantly, though, I don't think you should generalize from the need for traffic laws to the need for laws, period.

One can draw up a taxonomy of lawbreaking by how much the involved parties are motivated to avoid an illegal act. Basically, there are three types. (i) Both parties are motivated to avoid; (ii) one party is motivated to avoid, one is not; (iii) none of the parties are motivated to avoid. Traffic laws are of type (i). Sure, it may be to your advantage to speed ceteris paribus, but you don't want to get into an accident. Even if it's the other guy who's dead, your fender may be a writeoff. So, by and large, you're motivated to more or less reach some type of agreement with other traffic parties about who goes where when. These types of laws are hence not that important. Laws of type (iii) are especially hard to defend. A standard example is selling/buying drugs. Presumably, if both parties want to do it, it's because they're aiming to maximize their expected utility. Such laws are not impossible to defend, but if you want to do so, you have some work on your hands.

The biggie is type (ii) - one party is motivated to avoid, the other's not. Examples of type (ii) lawbreaking include homicide, rape, robbery, assault, burglary and theft. Interestingly, these are the "core crimes" which seem to be proscribed in pretty much all societies, including ones without written laws, as long as the victim's an ingroup member. (I've read conflicting claims about rape, though.) One possible explanation for this finding is that societies in which such behaviour was allowed once existed, but are long gone, because life was so nasty, brutish, and short that sooner or later no members were left. 

If you want to move towards anarchy, please do away with type (ii) laws last.